Project Summary Currently, drug embryotoxicity risk for safe pregnancy is not well established, thus many pregnant women are exposed to the drugs with unknown effects on fetus development. Many drugs are still neither well understood regarding their effects on human organogenesis, nor is there a well-established human embryotoxicity drug screening platform available. Currently, human induced pluripotent stem cells (hiPSCs) have been proposed for human-relevant drug toxicity screening. However, the use of hiPSC maintenance and differentiation on 2D culture is not an ideal embryotoxicity assay due to their inability to predict the drug toxicity on 3D tissue morphogenesis, which potentially leads to the structural malformations manifested in late prenatal fetus development. With the emergent concept of stem cell organoids, these 3D cultures of developing tissues imply the similarity to the manner in which different organs establish their characteristic organization during development. Therefore, the overall goal of this proposal is to establish an in vitro hiPSC-based cardiac organoid model for embryotoxicity testing based on the drug effects on hiPSC growth, cardiac differentiation, and early heart formation, so we can establish a risk classification system for more precise assessment of human-specific drug effects on early embryonic development. To achieve this goal, we will pursue three specific aims. In Aim 1, we will optimize the cardiac organoid model by investigating the effects of biophysical confinement on the formation and function of cardiac organoids. In Aim 2, we will validate the cardiac organoid-based embryotoxicity assay by comparing to well-established standard zebrafish whole embryo culture assay. By testing a ?training set? of chemicals with known embryotoxicity level, we will better calibrate the drug response from human cardiac organoids based on a variety of endpoint evaluation parameters. In Aim 3, we will establish a new biostatistical predictive model based on linear discriminant analysis for embryotoxicity risk classification. We envisage that this in vitro cardiac organoid model can improve traditional pharmaceutical screening for the drugs that will be administered during pregnancy.